IBM at its Think 2026 conference advanced a raft of artificial intelligence (AI) projects and initiatives while at the same time making a stronger case for investing in quantum computing platforms that will run certain classes of workloads faster than classical computers starting this year.
The latest AI advances from IBM include an update to the IBM watsonx Orchestrate that enables the platform to now also function as a control plane for managing AI agents. At the same time, IBM is making available in public preview a previously announced IBM Concert platform that makes use of AI agents to automate a range of IT operations.
IBM watsonx Orchestrate for multi-agent orchestration addresses a core need for a third party to handle cross-agent workflows across disparate vendors, says Keith Kirkpatrick, vice president and research director for the Futurum Group. “This is particularly important from both a governance and regulatory perspective,” he says’ “Getting all agents to work together in concert, without elevating or deprecating agents from a specific vendor or system, will be important in driving effectiveness and efficiency.”
IBM is also providing a public preview of HCP Terraform powered by Infragraph, a knowledge graph developed by its HashiCorp arm that integrates data across a distributed computing environment.
Additionally, IBM is making available in private preview IBM Context in watsonx.data that promises to make it simpler to apply semantic meaning, enforce governance at runtime, and make IT decisions explainable. Core capabilities include support for OpenRAG and OpenSearch on watsonx.data, and the Real-Time Context Engine from Confluent that IBM acquired earlier this year.
IBM is also providing a technical preview of a Presto capability in watsonx.data to accelerate query processing across graphics processing units (GPUs). In a proof of concept with Nestlé, IBM claims this engine delivered 83% cost savings and an overall 30x price-performance improvement on a global data mart spanning 186 countries. IBM is also making Confluent and Tableflow integration with watsonx.data now generally available.
There is now also in private preview an IBM Z Database Assistant that monitors performance, automates routine tasks, and optimizes configurations in mainframe environments.
IBM has added AI capabilities to IBM Vault 2.0 to discover secrets in code along with an IBM zSecure Secret Manager for mainframe environments that will be made available next month.
Finally, IBM also revealed that IBM Sovereign Core, a cloud-native platform that embeds policy at the infrastructure runtime level, is now generally available.
IBM chairman and CEO Arvind Krishna told conference attendees IBM is squarely focused on three primary vectors: AI-native infrastructure, hybrid clouds and quantum computing. In general, quantum computing is simultaneously being overhyped and under appreciated, said Krishna. This year will see these systems achieve a clear quantum advantage over classical systems, with IBM now having already deployed 80 quantum computers and now providing reference architectures to enable IT organizations to build and deploy applications, he added.
In the not too distant future, the output from quantum computers will also routinely be used to train AI models, he added. In fact, organizations such as the Cleveland Clinic are already advancing healthcare research by now being able to simulate a 12,635-atom protein using a quantum computer.
It’s not exactly clear what the ultimate impact the rise of AI and quantum computing will have, but the one thing that is certain is that as multiple computing architectures increasingly converge, there may very well be a new golden era of IT on the horizon.
